A Note on Panel Bias

Abstract

In analyzing continuous consumer panels for making marketing decisions, one must make some assumptions about how accurately the panel represents the total population of consumers. Yet very little work has been published on the representativeness of consumer panels. This article deals with a limited aspect of the problem: the effect of the length of time consumers have been in the panel on their purchasing characteristics. Tentative hypotheses are formulated on the basis of some empirical data and suggestions for future studies are given.